Today, I will look at the future of PLCs in the automation industry from a user's perspective. As a veteran in the automation field for over a decade, I must say that the industry is now incredibly diverse and vibrant. Why do I say this? Just as China is striving to become a manufacturing powerhouse with its "Made in China 2025" vision, Germany is launching its Industry 4.0 initiative, and the United States is raising the banner of the Industrial Internet, all automation manufacturers and industry veterans have proposed a common goal: the convergence of IT and OT. IT stands for Information Technology, and OT stands for Operational Technology (Automation Technology). These advanced technological concepts and forward-thinking solutions, emerging like mushrooms after rain, are relentlessly nourishing the Chinese people's thirst for knowledge about the integration of automation and information technology.
I hope automation users will approach this era of information overload with a rational mindset. Sit down calmly and think about what you truly need, and what automation products can offer you. Let's get to the point: what kind of PLC do you, as a user, really need? I don't want to reiterate the functions of a PLC here; everyone interacts with it directly or indirectly every day. You could say that PLC has become synonymous with the automation industry. Regarding the convergence of IT and OT, I'd like to offer a few points for consideration, regardless of which manufacturer's product it is.
1. High reliability of PLC.
Think about it: what has PLC been doing for the past 10 years? Constantly expanding its functions. Meanwhile, its stability has declined significantly. Many manufacturers are probably starting to prepare their bricks. I wonder how many people still remember Siemens' S5, Roche's PLC5, and Schneider's 984? Products from that era were extremely stable. Now, products often require G3 anti-corrosion treatment, but back then, there was no protection, yet it never failed. One PLC5 ran for 15 years in an aluminum workshop full of carbon dust, without ever needing spare parts replaced; they were simply too afraid to remove it for cleaning and maintenance. Why was that? Actually, the circuit boards of that era weren't highly integrated. PLC designers needed to perform parameter design and selection for the circuit boards and components. These designers were circuit enthusiasts back then, so the reliability and stability of PLCs were extremely high, though the price was also exorbitant. However, the user-friendliness was extremely poor, due to the limitations of operating systems at that time. But what about the reliability of today's PLCs? Thanks to the advent of highly integrated chips, today's PLC circuit boards are very simple. If you open one up, you'll see nothing but the chip and some basic peripheral circuitry. The advantage of this high integration is simplicity, but the downside is a significant reduction in reliability. Sometimes, manufacturers are helpless when it comes to EMC testing; the integrated chips simply have problems. What can be done? Users should prioritize the stability of the PLC itself. For large PLCs, especially those with redundant architectures, the primary task is to ensure stable operation of the control program, diligently collecting and outputting I/O logic before considering anything else.
2. The functionality of the PLC.
Next, let's talk about the functionality of PLCs. Westlife's TIA is booming, and the S1500 is thriving. The discussion is mostly about functionality, such as unified software tools (combining configuration and monitoring), unified control functions (motion control, drive control, process control), support for web-based diagnostics (a new feature of the S1500), and multiple network ports integrated on a single CPU (integrating the previous CP card). You can see that automation manufacturers have one goal: "integration"—to combine functions previously scattered across different product structures, and then proudly tell customers that they've developed a revolutionary intelligent PLC, the foundation for future industrialization and informatization integration. Seeing this makes me laugh; it's as if the lack of integrated functions would hinder the rapid progress of the automation industry. What users truly need is SDC (Software-Defined Controller), which means a unified and standardized programming and configuration tool. This allows for programming and configuring various PLCs through a single tool, whether for control or monitoring. Each PLC manufacturer only needs to provide the operating environment. Because the CPU chips used in PLCs are iterating rapidly, their performance and functionality are increasing exponentially. Just like ARM's ambition to challenge Intel, embedded CPUs are no longer just for show. So, I'd venture a guess: I personally believe that HMIs will disappear, replaced by web-based monitoring based on PLCbases. You can think of future PLCs as small computers; there's no task they can't do. At that point, functional limitations won't seem like an issue anymore.
3. The openness of PLC.
Here we mainly discuss networking, a topic I didn't initially intend to delve into, as it resonates with many manufacturers. Why? Everyone knows that the main battleground for automation over the next decade will be at the communication protocol level. Whether it's Profinet (S7), Ethernet/IP (CIP), EtherCAT, or Powerlink, they're all vying for market share. Industrial Ethernet organizations are everywhere, each eager to create their own GB/T standard, even aiming to eliminate GB/T altogether to gain a competitive advantage in automation. (Many will ask, "Have you forgotten Modbus TCP? It's been absorbed by ODVA, and its protocol stack has been integrated with CIP.") Then users will ask, which protocol is best? There's no inherent good or bad; it all depends on your choice. Industrial communication protocols fall into two camps: standard TCP/IP and those based on standard TCP/IP. (Ethernet/IP and PowerLink use the standard TCP/IP protocol stack, while Profinet and EtherCAT are based on TCP/IP.) These two are fundamentally different. As a customer, choosing the standard TCP/IP protocol means you can use industrial control products on a standard IT architecture without worrying about protocol conversion. You also don't need to worry about the future advancements of TCP/IP, as it's simply a layer 7 protocol built on top of it. But what if you choose a protocol based on TCP/IP? What does "based on" mean? It means you've modified the TCP/IP packets and even the layer 2 communication chip (EtherCAT). The advantage of this protocol is that it's specifically developed for industrial control applications. When building IT and OT convergence, edge devices need to use network devices supported by this protocol vendor. In basic control (IO, host computer, site-to-site), there's not much difference between the two, and users won't notice any difference. However, vendors using the standard TCP/IP protocol may have an inherent advantage in terms of openness. However, in specific industries and fields where extremely high communication speeds are required, the TCP/IP-based protocol still has its advantages. Just think about it! The original 150-byte TCP/IP message has been streamlined, which makes things much easier and more pleasant. In fact, it can even be sent out at the MAC layer!
